Silicon inhibition effects on the polymerase chain reaction:
A real-time detection approach
Wei Wang,1 Hai-Bin Wang,2 Zhi-Xin Li,1 Zeng-Yuan Guo1
1
Department of Engineering Mechanics, Tsinghua University, Beijing 100084, People’s Republic of China
2
The No. 302 Hospital, Beijing 100039, People’s Republic of China
Received 24 May 2005; revised 31 August 2005; accepted 31 August 2005
Published online 12 December 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.a.30627
Abstract: In the miniaturization of biochemical analysis
systems, biocompatibility of the microfabricated material is
a key feature to be considered. A clear insight into interactions between biological reagents and microchip materials
will help to build more robust functional bio-microelectromechanical systems (BioMEMS). In the present work, a realtime polymerase chain reaction (PCR) assay was used to
study the inhibition effects of silicon and native silicon oxide
particles on Hepatitis B Virus (HBV) DNA PCR amplification. Silicon nanoparticles with different surface oxides were
added into the PCR mixture to activate possible interactions
between the silicon-related materials and the PCR reagents.
Ratios of silicon nanoparticle surface area to PCR mixture
volume (surface to volume ratio) varied from 4.7 to 235.5
mm2/␮L. Using high speed centrifugation, the nanoparticles were pelleted to tube inner surfaces. Supernatant extracts were then used in subsequent PCR experiments. To
INTRODUCTION
Polymerase chain reaction (PCR), which is a thermally activated chemical reaction for nucleic acid amplification, is one of the most important molecular
biological methods.1 The number of nucleic acids will
be doubled after one thermal cycle by controlling the
sample and reagent temperature in sequence to the
denaturation temperature, the annealing temperature,
and the extension temperature with transitions. After
several tens of cycles, the amplified nucleic acids in the
PCR mixture will be rich enough for further detection
and analysis. It has been demonstrated experimentally
that the time spent in temperature transitions was
Correspondence to: W. Wang, Institute of Microelectronics,
Peking University, Beijing 100871, People’s Republic of
China; e-mail: [email protected]
Contract grant sponsor: Scientific Research Innovation
Foundation of Tsinghua University
Contract grant sponsor: National Natural Science Foundation of China; contract grant number: 59995550 –2
© 2005 Wiley Periodicals, Inc.
test whether silicon materials participated in amplifications
directly, in some cases, entire PCR mixture containing silicon nanoparticles were used in amplification. Fluorescence
histories of PCR amplifications indicated that with the increase in surface to volume ratio, amplification efficiency
decreased considerably, and within the studied ranges, the
higher the particle surface oxidation, the stronger the silicon
inhibition effects on PCR. Adsorption of Taq polymerase
(not nucleic acid) on the silicon-related material surface was
the primary cause of the inhibition phenomena and silicon
did not participate in the amplification process directly.
© 2005 Wiley Periodicals, Inc. J Biomed Mater Res 77A:
28 –34, 2006
Key words: real-time PCR; inhibition effects; silicon; nanoparticles
usually wasted and the transition time after the extension and the denaturation periods had no function.2
Furthermore, nonspecific amplification can be minimized with a rapid denaturation temperature to annealing temperature transition.3 Therefore, in the past
decade, development of a PCR instrument with rapid
temperature ramping, that is, rapid PCR, has become
an area of significant interest.
It is well known that characteristic time of thermal
transportation in a given device is direct proportional
to the second power of the device size. Therefore, a
thermal cycling instrument with characteristic size in
microscale is a perfect candidate to realize the rapid
PCR. With the development of silicon microfabrication techniques, silicon-based micro-PCR chips have
been widely studied to achieve rapid PCR as well as
reducing the power consumption and the required
space.4 –11 Silicon offers many substantial advantages
in establishing PCR instruments, such as very high
thermal conductivity and ease of fabrication. However, the use of silicon in micro-PCR chips poses some
drawbacks, especially in terms of the biocompatibility
issue. The high surface to volume ratio increases the
SILICON INHIBITION EFFECTS ON PCR
importance of the surface chemistry on the biological
operations in these micro devices.12
Native silicon was found to be an inhibitor of PCR
and nucleic acids amplification in an untreated siliconbased PCR chip had a high failure rate.13 To render the
inner surface of silicon-based micro-PCR chips, surface passivation procedures have been developed,
which can be classified into two types: static passivation and dynamic passivation.14 Static passivation
means that before performing PCR, the surface is
treated by oxidizing silicon4,6,9 –11 or by silanizations,5,8 while the dynamic approach refers to adding
the passivating agents directly into the PCR mixture.7,8,10,14 Early studies revealed that passivating the
silicon surfaces with a silanizing agent followed by a
polymer treatment can result in a good amplification
but the yields were inconsistent within the different
treatments and were not always comparable with PCR
in the conventional tubes.13 Further research indicated
that silanization of SiOx surfaces alone will not be
suitable for multiple PCR or long-term application
because of degradation of the surface-passivating organic film.15 Most dynamic coatings performed for
silicon-based PCR chips have used bovine serum albumin (BSA) as the passivating agent. It has been
found that in conventional amplification experiments,
high concentrations of BSA caused lower yields than
controls (without BSA added) and amplification could
even be shut down at a very high BSA concentration.6
Hence, a given PCR reaction in silicon-based microchip should be optimized in advance to find the optimal BSA concentration. Among the passivating methods, thermally oxidized silicon surfaces were
demonstrated to be compatible with PCR to give consistent amplifications when compared with reactions
performed in conventional PCR tubes.13
As for the inhibition mechanism, more and more
data suggest that the inhibition of PCR by siliconrelated materials is mainly due to the adsorption of the
Taq polymerase and/or the nucleic acid onto the chip
inner surfaces and not from a straight chemical action.6,15,16
In experiments reported for analysis of silicon inhibition effects on PCR, the PCR amplification results
were all detected by gel electrophoreses.13–16 This offchip manual operation increases the potential for cross
contamination, and the additional post-PCR step can
only provide a qualitative insight into the inhibition
phenomena. This implies that the different gel band
fluorescence intensities in the electrophoresis cannot
be simply taken as evidence of inhibition effects. In the
recent past, real-time PCR-based assay has become an
invaluable tool for many scientists working in different disciplines, especially in the field of molecular
diagnostics. The real-time PCR technique was characterized by a wide dynamic range of quantification, a
29
high sensitivity and high precision.17,18 No post-PCR
steps required minimizes risks of cross contamination.
The aim of the present work was to provide a quantitative insight into silicon substrate inhibition effects
on polymerase chain reaction by a real time PCR
approach. Two kinds of silicon nanoparticles with
different oxidation states at their surfaces were used as
experimental silicon materials. Reactions between the
silicon-related materials and the PCR mixture were
evoked by mixing the silicon nanoparticles with the
PCR reagents, and subsequently stopped by a high
speed centrifugation to sedimentate the nanoparticles
onto the inner surface of the experimental tubes. Fluorescence histories of the real-time detection were
used to evaluate the inhibition effect and to explore its
mechanism.
MATERIALS AND METHODS
Silicon particulate materials
To minimize the parasitic thermal mass introduced by the
additional silicon materials, sol– gel prepared silicon nanoparticles with a nominal diameter ranging from 20 to 50 nm
(ZhongChao Nano, China) have been used as the test materials in the present experiments. Two kinds of silicon nanoparticles were examined: (1) silicon nanoparticle exposed in
the air for a long time with a native oxide layer formed on its
surface, named as silicon nanopaticle I hereafter, and (2)
nonoxidized silicon nanoparticles which were taken out
from a nitrogen protected bottle just before experiments,
named as silicon nanoparticle II.
X-ray photoelectron spectroscopy (XPS, PHI-5300 ESCATM, PerkinElmer, USA) was used to evaluate the surface
oxidation degree of the two nanoparticles by measuring the
spectra of the two nanoparticles, compared with a silicon
wafer with a thick thermal oxide layer (fabricated by standard industrial thermal oxidation process in Hebei Semiconductor Research Institute, Shijiazhuang, China; thickness of
its oxide layer was larger than 110 nm). An aluminium
anode was used as source, operating at 250 W. The binding
energies were lined up with respect to the C1s peak at 285
eV. Wide-scan spectra for the three samples are illustrated in
Figure 1(A). Si2p high resolution spectra for the three samples, as shown in Figure 1(B), indicate a clear Si2p peak at
103.3 eV in the nanoparticle-I spectrum, signifing that the
nanopartilce-I was slightly oxidized. Mean while, XPS of
nanoparticle-II exhibits a very weak Si2p peak at 103.3 eV,
but a distinct one at 99.3 eV. No Si2p peak at 99.3 eV for the
thermal oxide SiO2 sample was detected because the thickness of its oxide layer was much larger than the inelastic
mean free path of XPS. Thickness of the silicon oxide overlayer can be calculated from the spectrum by using the
following equation19
再
tox ⫽ ␭SiO2 sin ␪
冎
exp
ISiO
/ISiexp
2
⫹1
␤
(1)
where ␪ was the angle between the sample surface plane and
the electron analyzer, ␭SiO2 was the attenuation length of the
30
WANG ET AL.
nanoparticle-suspended liquid was 186 ⫾ 0.8 mm2/␮L.
Mastersizer 2000TM (Malvern Instruments, UK) was used to
evaluate the cluster degree of the two kinds of nanoparticlesuspended liquids. Results showed that the average diameters of nanoparticle I and II clusters in sterile water were 399
and 297 nm, respectively.
PCR Protocol
Figure 1. XPS spectra for the two silicon nanoparticles with
a thermal oxide silicon wafer. (A) Wide-scan spectra for the
three samples, (B) Si2p high resolution spectra for the three
samples (selected data from panel A).
Si2p photoelectrons in SiO2, and ␤ was the ratio of the Si2p
intensity from infinitely thick SiO2 to that of Si. In the
present experiments, ␪ was 45°, and as recommended by
Shallenberger et al 20 ␭SiO2 and ␤ were taken as 2.7 nm and
0.83, respectively. Therefore, the thickness of the native silicon oxide films of the nanoparticle I and II were calculated
to be about 3.33 and 2.52 nm, respectively.
The specific surface area (surface area per unit mass) of
the two kinds of nanoparicles were measured by Brunauer–
Emmet–Teller adsorption method (BET, NOVA4000TM,
Quantachrome, USA). On the basis of nitrogen adsorption,
the specific surface area of the silicon nanoparticle I was
47.63 m2/g and that of the nanoparticle II was 77.83 m2/g.
To guarantee that the silicon-related material surface areas
in the individual experimental tubes were the same; 3.9 mg
of silicon nanoparticle I and 2.4 mg silicon nanoparticle II
were added into two tubes with 1 mL sterile water, respectively. The ratio of silicon surface area to volume of each
In the present experiments, the amplification target was
DNA of Hepatitis B Virus (HBV, clinical samples provided
by No. 302 Hospital), and the buffer in the present experiment contained 10 mM Tris–HCl, pH 8.3, 50 mM KCl, 3.5
mM MgCl2, 400 mM dUTP, 200 mM each dATP, dCTP, and
dGTP, 300 nM each primer, and 200 nM probe (Shanghai
Sangon Co., China). The sequence for the forward primer
was 5⬘-ATC CTG CTG CTA TGC CTC ATC TT-3⬘, and the
sequence for the reverse one was 5⬘-ACA GTG GGG GAA
AGC CCT ACG AA-3⬘ (Shanghai Sangon Co., China). Together, these two primers define a 103 bp PCR product. For
homogeneous detection using the TaqManTM system, a fluorescent probe was also included. The sequence of the HBVDNA fluorescent probe was 5⬘-R-TGG CTA GTT TAC AGT
GCC ATT TG-Q-3⬘ (Shanghai Sangon Co., China), where R
was the fluorescent reporter dye and Q was the fluorescent
quencher dye.
In ilka amplification, the PCR mixture containing 35 ␮L
buffer, 0.5 ␮L Taq polymerase (1.25 U/␮L AmpliTaq Gold
DNA polymerase, PerkinElmer), and 4 ␮L DNA template
(extracted and purified from the clinical sample by QIAamp
DNA Blood Mini KitTM, Qiagen, USA) was mixed with the
silicon nanoparticles (volume ignored) as described in the
following section. The pre-denaturation period was 95°C for
120 s, and then 35 cycles of amplification were performed
using a thermal profile of 94°C for 20 s and 53°C for 30 s. The
amplifications were carried out in a real-time PCR instrument, SlanTM (Hong Shi, China).
Experimental procedures
To evaluate the silicon inhibition effects on HBV DNA amplification, an experiment was designed, which is referred as
“inhibition experiment” hereafter. First, the two silicon nanoparticle-suspended liquids were oscillated to homogeneity.
Then, 1, 2, 5, 10, 20, and 50 ␮L of the two liquids was distributed into 12 tubes, respectively. A 12,000 rpm (about 1000⫻g)
rotation for 3 min made the silicon nanoparticles partition to
the inner surfaces in each tube. Supernatant was aspirated. The
PCR mixture prepared beforehand was added into the 12
tubes, and then oscillated with the pelleted silicon nanoparticles to regenerate a homogeneous nanoparticles suspension.
The liquid was then allowed to stand for 3 min to make the
reagents contact the silicon nanoparticles adequately. Another
3-min 12,000 rpm rotation was used to separate the nanoparticles from the PCR mixture and was treated as a stopping
“reagent” of the possible reaction between the PCR reagents
and the silicon-related materials. The supernatants were extracted again and injected into another 12 fresh tubes. These 12
SILICON INHIBITION EFFECTS ON PCR
31
TABLE I
PCR Mixture Preparation in Mechanism Experiments
Silicon nanoparticles
Adsorption objects
Adding reagent
Silicon remaining
Tube 1
Tube 2
Table 3
Tube 4
I
Enzyme ⫹ buffer
Template
Yes
I
Enzyme ⫹ buffer
Template
No
I
Template ⫹ buffer
enzyme
Yes
I
Template ⫹ buffer
enzyme
No
test samples with 2 positive controls (without silicon nanoparticles) and 1 negative control (without the HBV DNA template)
were thermally cycled in the SlanTM real-time PCR instrument.
Another experiment (“mechanism experiment”) was designed and carried out to distinguish the reactive reagent in the
PCR mixture with the two kinds of silicon nanoparticles and to
test whether the silicon material participated in the nucleic acid
amplification directly. First, prepared nanoparticle I suspended
samples were oscillated to homogeneity and dispersed into 4
tubes, each tube with 50 ␮L volume. Rotation for 3 minutes at
12,000 rpm was used to pellet the nanoparticles to the tube
inner surface, and then supernatants were extracted from the
tubes. Template– buffer or enzyme– buffer mixtures were
added into the different numbered tubes as illustrated in Table
I. The added reagents and sedimented nanoparticles were oscillated to homogeneity again, and stabilized for 20 min to
allow the reaction between the silicon materials and the PCR
mixture progress completely. Another 12,000 rpm rotation pelleted the silicon nanoparticles to the tube inner surfaces, and
then the absent reagent (template or enzyme) was injected into
each tube. Supernatants in Tubes 2 and 4 were extracted and
injected into two fresh tubes, respectively, and the following
amplifications were performed without silicon nanoparticles.
The other two tubes were placed into the SlanTM PCR instrument directly with the nanoparticles inside for PCR amplification. The parasitic thermal mass effect introduced by the additional silicon nanoparticles was neglected because the thermal
mass of silicon nanoparticles inside tube was only about 0.19%
of that of the PCR mixture. The four test tubes were amplified
simultaneously with 2 positive controls (without silicon nanoparticles) and 1 negative control (without the HBV DNA template) in the SlanTM real-time PCR instrument.
Assay repeatability was tested and the results indicated
that the SlanTM real-time PCR instrument can provide a
robust and reliable nucleic acid amplification (data not
shown).
sulted in an amplification efficiency reduction. And with
the increase in the particle surface to volume ratio, the
fluorescence intensities in amplification decreased considerably. As shown in Figure 2(C), in the experiments
using silicon nanoparticle I (surface to volume ratio of
235.5 mm2/␮L), fluorescence intensity in the tube at the
35th cycle was about 66% of that of the positive control,
while in the experiments with silicon nanoparticle II, the
factor was about 83%. Results indicate that the higher
oxidized silicon surface (native oxide on silicon nanoparticle I) exhibited a much stronger inhibition effect on the
HBV DNA amplification. This conclusion seemed inconsistent with previous results 14 –16, claiming that the silicon oxide was compatible with PCR but pure unoxidized silicon was not.
These apparent contradicting phenomena can be
attributed to variations of silanol (active site) densities
at the surface with different oxidation degrees and
thermal treatment methods. Silanol surface was reported to be able to adsorb proteins considerably,21
and Taq polymerase adsorption on surface was proposed to be the primary cause for silicon inhibition
effects on PCR as discussed in the next section. For
naked silicon, the surface has a low oxygen concentration and presents a very small number of hydroxyl
sites, while the natural oxidation produces a SiOO
surface, which is easily oxidized to silanols in aqueous
solution. Preparations of PCR-compatible silicon oxides in the previous works usually include a very high
temperature thermal treatment (ca. 1000°C). This high
temperature oxidization process makes silicon surface
in the form of OOSiOO and stable against the hydrolysis processes.22 Therefore, the hydroxide (silanol) is
difficult to form on the surface.
RESULTS AND DISCUSSION
Inhibition mechanism
PCR inhibition performance
In the inhibition experiment, fluorescence histories in
the 15 experimental tubes (12 samples, 2 positive controls, and 1 negative control) for tubes containing silicon
nanoparticle I and silicon nanoparticle II were shown in
Figure 2. The real-time fluorescence records indicated
that regardless of the oxidation degree of the nanoparticles surfaces, the existence of silicon-related materials
in the PCR mixture preparation procedure always re-
Fluorescence histories comparing the amplifications
with DNA adsorption tests and Taq polymerase adsorption tests in the mechanism experiments with silicon nanoparticle I (surface to volume ratio of 235.5
mm2/␮L) are shown in Figure 3(A). Amplification
efficiencies decreased considerably in the Taq polymerase adsorption test, and fluorescence intensity in
the tube at the 35th cycle was about 64% of that of the
positive control, while in the DNA adsorption test, the
32
WANG ET AL.
Figure 2. Fluorescence histories of HBV-DNA amplifications with different surface to volume ratios in inhibition experiment
(with 2 positive controls and 1 negative control). (A) Results for silicon nanoparticle I, (B) results for silicon nanoparticle II,
and (C) comparison of fluorescence histories of silicon nanoparticle I and II with a surface to volume ratio of 235.5 mm2/␮L
(selected data from panels A and B).
SILICON INHIBITION EFFECTS ON PCR
33
Figure 3. Fluorescence histories of the HBV-DNA amplifications in mechanism experiment (with 2 positive controls and 1
negative control). (A) Results of amplifications in DNA adsorption tests and Taq polymerase adsorption tests with silicon
nanoparticle I (surface to volume ratio of 235.5 mm2/␮L). (B) Results of amplifications with and without silicon nanoparticles
(silicon nanoparticle I with a surface to volume ratio of 235.5 mm2/␮L) in Taq polymerase adsorption tests.
factor was about 92%. Ct values, the threshold cycle
number at which fluorescence begins to increase rapidly, usually a few standard deviations above the
baseline, in the both tests did not shift, as illustrated in
Figure 3(A). Because the plot of Ct value versus template number is linear in the real time PCR assay, the
observed non Ct value-shift means that there is no
detectable DNA adsorption on the silicon nanoparticles. Results demonstrated that the observed inhibition effects can be attributed to the adsorption of the
Taq polymerase on the silicon oxide surface and that
the adsorption of the template (DNA) is not a primary
factor. Amplification results for nanoparticle II are the
same as those for nanoparticles I (data not shown),
except that the intensities of the inhibition are different. This is also the same as those found in gel electrophoresis detection.23
By controlling the silicon nanoparticles remaining in
the thermal cycling process, the straight inhibition
phenomena on PCR was tested. Fluorescence histories
of PCR amplifications with and without silicon nanoparticles (silicon nanoparticle I, surface to volume ratio of 235.5 mm2/␮L) in the Taq polymerase adsorption tests are shown in Figure 3(B). It is clear that
under these two kinds of conditions, the nucleic acid
amplification efficiencies remain nearly the same. Similarly, amplifications with and without silicon nanoparticles in the DNA adsorption tests have the same
efficiencies (Data not shown).
We conclude that silicon nanoparticles have no direct inhibition effects on the PCR amplification and the
inhibition phenomena should be ascribed to the adsorption of the Taq polymerase on the silicon related
material surface.
34
WANG ET AL.
CONCLUSIONS
5.
As a mature fabrication material used in the MEMS,
silicon has been widely used to fabricate miniaturized
PCR devices. However, in the silicon-based microPCR chips, surface chemistry plays an important role
in nucleic acid amplification. We have shown that
silicon produces a detectable inhibition on the PCR
reaction, confirming previous works.
In the present work, a real-time polymerase chain
reaction assay was used to provide a quantitative
insight into this inhibition phenomenon. In the experiments, silicon based nanoparticles with high specific
areas were mixed with the PCR mixture to evoke
interaction between them. A broad range of surface
analysis techniques including X-ray photoelectron
spectroscopy and Brunauer–Emmet–Teller adsorption
isotherms were used to characterize the two nanoparticles before the experiments. In PCR experiments,
ratios of the silicon/silicon oxide surface area to the
PCR mixture volume varied from 4.71 to 235.5 mm2/
␮L. Fluorescence intensities detected by the real-time
PCR instrument indicated that with the increase in
surface to volume ratio, the amplification efficiency
decreased considerably, and particles with native oxide exhibited a much greater inhibition effect on nucleic acid amplification than naked silicon particles.
Enhanced PCR inhibition phenomena were attributed
to the very large number of surface silanol groups
present on the native oxide surface. Experimental results suggested that the adsorption of Taq polymerase
(not nucleic acid) onto the silicon surface was the
primary cause of the PCR inhibition and silicon did
not participate in the amplification process.
The next phase of this work will scrutinize the inhibition effects of other silicon-related materials, including silicon nitride, silicon dioxide (wet oxidation,
dry oxidation, and chemical vapor deposited oxidation), polysilicon, and other materials. Further studies
on the adsorption of Taq-polymerase on the microfabricated surface will also be a focus.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
References
1.
2.
3.
4.
Mulli KB, Ferré F, Gibbs RA. The Polymerase Chain Reaction,
Birkhäuser: Boston; 1994.
Wittwer CT, Fillmore GC, Garling DJ. Minimizing the time
required for DNA amplification by efficient heat transfer to
small samples. Anal Biochem 1990;186:328 –331.
Wittwer CT, Garling DJ. Rapid Cycle DNA amplication: Time
and temperature optimization. Biotechniques 1991;10:76 – 83.
Cheng J, Shoffner MA, Hvichia GE, Kricka LJ, Wilding P. Investigation of different PCR amplification systems in microfabricated
silicon-glass chips. Nucleic Acids Res 1996;24:380 –385.
21.
22.
23.
Poser S, Schulz T, Dillner U, Baier V, Köhler JM, Schimkat D,
Mayer G, Siebert A. Chip elements for fast thermalcycling,
Sensor Actuators A 1997;62:672– 675.
Taylor TB, Winn-Deen ES, Picozza E, Woudenberg TM, Albin
M. Optimization of the performance of the polymerase chain
reaction in silicon-based microstructure. Nucleic Acids Res
1997;25:3164 –3168.
Daniel JH, Iqbal S, Millington RB, Moore DF, Lowe CR, Leslie
DL, Lee MA Pearce MJ. Silicon DNA microchambers for DNA
amplification. Sensor Actuators A 1998;71:81– 88.
Schneega␤ I, Bräutigam R, Köhler JM. Miniaturized flowthrough PCR with different template types in a silicon chip
thermocycler. Lab on a Chip 2001;1:42– 49.
Lee DS, Park SH, Yang H, Chung KH, Yoon TH, Kim SJ, Kim
K, Kim YT. Bulk-micromachined submicroliter-volume PCR
chip with very rapid thermal response and low power consumption. Lab on a Chip 2004;4:401– 407.
Matsubara V, Kerman H, Kobayashi M, Yamamura S, Morita
V, Takamura Y, Tamiya E. On-chip nanoliter-volume multiplex TaqMan polymerase chain reaction from a single copy
based on counting fluorescence released microchambers. Anal
Chem 2004;76:6434 – 6439.
Krishnan M, Burke DT, Burns MA. Polymerase chain reaction
in high surface-to-volume ratio SiO2 microstructures. Anal
Chem 2004;76:6588 – 6593.
de Mello AJ. DNA amplification: does ‘small’ really mean
‘efficient’? Lab on a Chip 2001;1:24N–29N.
Shoffner MA, Cheng J, Hvichia GE, Kricka LJ, Wilding P. Chip
PCR. I. Surface passivation of microfabricated silicon-glass
chips for PCR, Nucleic Acids Res 1996;24:375–384.
Lou XJ, Panaro NJ, Wilding P, Fortina P, Kricka LJ. Increased
amplification efficiency of microchip-based PCR by dynamic
surface passivation. Biotechniques 2004;36:248 –252.
Felbel J, Bieber I, Pipper J, Köhler JM. Investigations on the
compatibility of chemically oxidized silicon (SiOx)-surface for
applications towards chip-based polymerase chain reaction.
Chem Eng J 2004;101:333–338.
Erill I, Campoy S, Erill N, Barbé J, Aguiló J. Biochemical
analysis and optimization of inhibition and adsorption phenomena in glass-silicon PCR-chips. Sensor Actuators B 2003;
96:685– 692.
Schmittgen TD. Real-time quantitative PCR, Methods 2001;25:
383–385.
Klein D. Quantification using real-time PCR technology: Applications and limitations, Trends Mol Med 2002;8:257–260.
Hill JM, Royce DG, Fadley CS, Wagner LF, Grunthaner FJ. Properties of oxidized silicon determined by angle-dependent X-ray
photoelectron spectroscopy. Chem Phys Lett 1976;44:225.
Cole DA, Shallenberger JR, Novak SW, Moore RL, Edgell MJ,
Smith SP, Hitzman CJ, Kirchhoff JF, Principe E, Nieveen W,
Huang FK, Biswas S, Bleiler RJ, Jones K. Oxide thickness
determination by XPS, AES, SIMS, RBS and TEM. J Vac Sci
Technol B 2000;18:440 – 444.
Alcantar NA, Aydil ES, Israelachvili JN. Polyethylene glycolcoated biocompatible surface, J Biomed Mater Res 2000;51:343–
351.
Dugas V, Chevalier Y. Surface hydroxylation and silane grafting on fumed and thermal silica. J Colloid Interface Sci 2003;
264:354 –361.
Wang W, Wang HB, Li ZX, Guo ZY. Studies of silicon inhibition effects on polymerase chain reaction, China Biotechnology
2005;25(3):69 –73.